Unordered Landmark Visual Navigation

2026-08-07

AI Summary

This work addresses the challenges of perceptual ambiguity and mapping failure in visual navigation when relying solely on unordered RGB images. It proposes the first unified framework that operates without temporal sequences or auxiliary sensors such as depth cameras or LiDAR. The method constructs a 2D topological map through geometric verification and maximum spanning forest formation, and integrates graph neural reasoning with an entropy-adaptive belief propagation mechanism to jointly perform mapping, global localization, and dynamic sub-goal planning. Experimental results demonstrate that the approach significantly outperforms state-of-the-art methods in both simulated and real-world environments, establishing—for the first time—the feasibility of robust and scalable visual navigation under completely unordered image conditions.

Abstract

Image-goal navigation is a fundamental capability for embodied AI, yet its practical deployment is strained by strong prior assumptions. Existing methods predominantly rely on temporally ordered video streams or auxiliary sensors (e.g., depth, LiDAR) to maintain spatial consistency. These sequential and multimodal dependencies severely restrict scalability, especially when deploying robots using crowd-sourced or pre-recorded unordered image collections. When temporal priors are removed, current methods struggle with severe perceptual aliasing, noisy associations, and catastrophic mapping failures. To address this underexplored challenge, we propose Unordered Landmark Visual Navigation (ULVN), a unified RGB-only framework free from temporal and odometric priors. ULVN systematically mitigates error accumulation by integrating mapping, localization, and planning. Specifically, it constructs a robust 2D topological map directly from unstructured images via calibrated geometric verification and maximum spanning forest refinement. For closed-loop execution, ULVN abandons sequential heuristics, utilizing a graph-based belief propagation filter with entropy-adaptive fusion for global localization and dynamic subgoal planning. Extensive experiments in simulation and real-world deployments demonstrate that ULVN significantly outperforms state-of-the-art methods.

Problem

• image-goal navigation
• unordered images
• perceptual aliasing
• spatial consistency
• embodied AI

Innovation

• unordered visual navigation
• RGB-only navigation
• topological mapping
• graph-based belief propagation
• entropy-adaptive fusion